* rsalveti back00:03
rsalvetisveinse: you can just recompile qt with mesa, using gles and then use the sgx libraries00:05
rsalvetithe current sgx package doesn't provide the headers needed to build packages, because of the lack of a proper soname00:06
rsalvetiso if you plan to build and use anything with gles, just do it with the mesa libraries and then install the sgx when you want to test with it00:06
topfs2sooo annoying though :)00:14
topfs2guess crosscompiling would help though00:14
=== DanaG_ is now known as DanaG
* sveinse back00:17
sveinseOK. here I am building both the SGX package and qt from sources, both cross compiled00:18
sveinsersalveti: what is mesa, and how does it fit with Qt. I mean is it any different from compiling Qt and SGX together. (I have very little knowledge of what is inbetween these two sets of software)00:19
rsalvetisveinse: why are you cross compiling the sgx package?00:20
rsalvetiit's just binaries00:20
sveinsePartly history00:20
rsalvetithe only thing to compile there is the kernel modules, and that doesn't help your qt build00:20
rsalvetisveinse: mesa is the software implementation of gl and gles00:21
rsalvetisgx should just replace it fine, but we don't have a proper soname at the sgx libraries00:21
sveinseWe're building a target system of Ubuntu Maverick. Since we're this early in the development, we haven't had time to setup proper .deb package generation00:21
sveinseSo we have some software being compiled, custom kernel (for custom board), custom Qt (since we're not using X11, but want GLES), and so on00:22
rsalvetioh, ok00:22
sveinseAnd. we want to cross compile, since not able to do so make our build server useless00:22
sveinsePlan is though, that we migrate over to standard debs later on (since we want to be able to apt-get update our system).00:23
rsalvetiok, now I get it00:23
sveinseUsing the powervr-omap3-dkms is something I want to do00:24
sveinseFYI: I have been able to make a system for cross building apps into a ubuntu target system without too many worries.00:25
sveinseThanks to the armel-cross package!00:25
sveinseHave to say I'm a little proud of it, but noone here really understands that ;)00:25
sveinseComing back to the issue here: I'm looking at compiling SGX (or repackageing it, since its mostly bins) and compiling Qt with support for it00:26
sveinseYou say I can compile Qt with generic mesa support, you say?00:27
rsalvetiok, so the only thing currently missing are the khronos headers00:27
rsalvetiyou can build with the sgx libraries, if you grab the correct khronos headers, that you can already find at the mesa-dev packages00:27
rsalvetiI believe it'll be easier for you now if you just build agains the sgx libraries00:28
rsalvetias you're not worried about proper packaging atm00:28
sveinseatm, yes. But it sounds like this will backfire later...00:28
rsalvetithe only problem is that the lack of a proper soname makes it incompatible with the mesa packages00:29
rsalvetiwe, as a distro, cares about that because we can't be sure that the user will install the drivers, and that the board will contain a sgx chip00:29
sveinseI remember something about some headers while compiling Qt some while ago during the concept test... Caused by the missing Khronos headers, right00:29
rsalvetiso we need the generic implementation to work00:29
sveinseAh, since they are hardcoded in the SGX package00:30
rsalvetiyes, so currently on ubuntu if we build something against mesa, it works with the sgx package00:30
rsalvetibut not the opposite00:30
rsalvetiso if you only care about using your software on omap with sgx installed, it's fine for you to just build agains the sgx packages00:31
sveinseyes, I do, thanks00:31
rsalvetithe only thing you'll need atm is the proper header files, as we currently don't deploy it00:31
sveinseIs the missing headers a bug/glitch, or is it an NDA thing?00:32
rsalvetiit's just because we don't want people building packages with the sgx one at ubuntu in general00:33
rsalvetito avoid bugs with the lack of a proper soname00:33
rsalvetibut you can find the headers at the mesa package00:34
rsalvetilet me look it for you00:34
sveinseI admit I have to check google what mesa is...00:34
rsalvetisveinse: it's an opensource implementation of gl and gles00:35
rsalvetiand you can run gl/gles by software if you need00:35
rsalvetieven on your desktop you can now run gles applications with mesa00:35
rsalvetiyou can also have drivers to accelerate it00:35
sveinseInteresting thing though. The first word in OpenGL is open....00:37
rsalvetisveinse: but an open implementation :-)00:38
rsalvetisveinse: check http://ports.ubuntu.com/ubuntu-ports/pool/main/m/mesa/00:38
rsalvetipackages: libegl1-mesa-dev libgles1-mesa-dev libgles2-mesa-dev00:38
rsalvetithere you'll find all the needed headers to build for gles00:38
rsalvetior just grab the mesa source and look for the files00:39
rsalvetisveinse: http://paste.ubuntu.com/543851/00:39
* sveinse is still exited to see what effect GL has on the speed and resposiveness of his Qt app...00:40
sveinse<wishlist>If only Ubuntu had Qt compiled for qws without X11, I wouldn't have to recompile Qt</wishlist>00:42
rsalvetisoon we'll have arm public ppas and you'll be able to create a ppa for that :-)00:42
sveinsesorry, ppa?00:43
ScottKsveinse: If only Qt upstream provided a supported way to do that without having to maintain two complete copies of Qt.00:44
sveinseYeah, I see that. To my knowledge it is only libQtGui.so which is bound to the gfx-system, and the rest isnt. But you still need to make that distinction while configuring00:45
rsalvetisveinse: ppa is one way to have a public repository at the launchpad, so you can submit your packages, launchpad will  build and make them available for you00:46
rsalvetithen you can use the repository entry at your sources.list00:46
rsalvetiScottK: that would be nice to see, but don't think they will put resources on it00:47
ScottKrsalveti: That's the reason there's no qt-qws in the archive.  It's a lot of maintenance work/buildd time to double it.00:48
sveinseHaving worked with setting up a proper build system for cross compiled Qt apps, I have a handful of comment about the Qt build system....00:48
ScottKProbably none of the printable in public ....00:48
sveinseWell (without starting another fight) its similar to Mac's: If you fit inside the box Apple and/or Nokia has prepared for you it's great! If you dont: Well then the box become Pandora's box...00:50
sveinsersalveti: FYI: You know that TI has released a version of their SGX?00:52
rsalvetisveinse: yes, but afaik the only thing they changed was the support for one additional chip00:54
rsalvetithat's why we didn't update the package00:54
sveinsegreat, thought you should know00:54
rsalvetithanks anyway :-)00:54
sveinseI think I will use the powervr dkms driver for the kernel part, and my custom Qt+SGX, since I need to custom build Qt anyways00:55
sveinseDid you come to a solution in regards of the devmem2 thing?00:56
sveinseSeems like I have to make something in upstart. The SGX driver wants to do some magic during bootup00:57
sveinseMagic seems to be pvrsrvinit, of which is a precompiled binary thing00:59
sveinseWell, I'm really going to bed this time01:00
sveinsersalveti: Thanks a lot for you assistance01:00
sveinseAppreciate it!01:00
rsalvetisveinse: yes, check the upstart file from our sgx package01:01
rsalvetiyou'll see the hack heh :-)01:01
rsalvetiyou first need to identify the omap 3 version and then use the proper pvrsrvinit01:01
rsalvetinp, ping me later if you have issues with it01:02
sveinseThe powervr-omap3 package?01:02
rsalvetisveinse: opengles-sgx-omap301:03
sveinseHm.. Not available from my source (http://ports.ubuntu.com/ubuntu-ports maverick main universe)01:04
rsalvetitry multiverse01:05
sveinsegot it01:06
sveinseWell, thanks. Going to look at it in the morning. Night01:08
=== asac_ is now known as asac
=== Baybal__ is now known as Baybal
guerbyrsalveti, I added a bunch of info to https://bugs.launchpad.net/ubuntu/+source/linux-ti-omap4/+bug/69037007:44
ubot2Launchpad bug 690370 in linux-ti-omap4 (Ubuntu) "Strange out of memory on pandaboard (affects: 1) (heat: 8)" [Undecided,New]07:44
guerbyrsalveti, I can provide you ssh root access to my pandaboard07:44
guerbyrsalveti, I sent you access instructions07:58
guerbyndec, same :)07:58
ndecguerby: cool. i am connected ;-)08:26
guerbyndec, I'll be mostly offline today, synchronize with rsalveti for what you do on it :)08:27
guerbythe processes running are unimportant08:27
ndecguerby: i don't have much to do on it. i have several panda already ;-)08:28
guerbyndec, I don't know what's so special with mine to get all those mem errors08:28
ndecguerby: well, i think i would get the same errors on mine if i was doing the same things as you are.08:29
ndecguerby: that said, we use the panda pretty heavily since we do a lots of native builds...08:29
guerbyndec, do you use NFS?08:30
ndecguerby: not me, but some other folks around do. i work mainly on SD or on USB (host)08:30
guerbyndec, ok, we'll see what you find out :) If it's too much trouble I'll switch to USB disk for /home or / as documented on omappedia08:34
alf_rsalveti: Hi! Are there any known issues regarding the XDamage extension on OMAP4?10:20
rsalvetialf_: not that I know, why?13:08
rsalvetiguerby: hey, will look at the bug13:10
rsalvetijust need to wake up first :-)13:10
alf_rsalveti: I am trying to make a small program using EGLImage + XComposite + XDamage and I am getting some problems with Damage events not being reported on some windows when redirecting the whole hierarchy13:10
alf_rsalveti: if I redirect individual windows I don't have a problem13:11
alf_rsalveti: it may be a bug in my code... but then again it works on my desktop13:11
rsalvetihm, ok13:11
rsalvetialf_: can you reproduce with the fbdev x11 driver?13:12
rsalvetiby just using xcomposite and xdamage?13:12
alf_rsalveti: haven't tried yet... will do and let you know13:12
alf_rsalveti: thanks13:13
rsalvetialf_: ok, if it works it's probably a bug at the pvr x11 driver13:13
=== njpatel is now known as njpatel_
ScottKDoes Natty support -mthumb-interwork?14:00
=== ian_brasil__ is now known as ian_brasil
guerbyrsalveti, hi :) first bootstrap cycle succeeded: http://gcc.gnu.org/ml/gcc-testresults/2010-12/msg01284.html14:13
hrwScottK: iirc yes14:16
ScottKhrw: Thanks.14:17
rsalvetiguerby: nice, so that's with 768 and swap?14:19
=== sbambrough is now known as scottb-afk
guerbyrsalveti, yes. It still gets backtraces in messages but it had no impact I can see on the GCC bookstrap+check14:20
=== zyga is now known as zyga-food
rsalvetiguerby: cool, good to know14:21
guerbyrsalveti, with swap off I got user program fail because of malloc returning NULL but no backtrace at all in messages (with mem hole or not)14:21
rsalvetiguerby: ok14:21
rsalvetiwill later try to debug with a different vm split and without highmem14:22
guerbyrsalveti, you should be able to ssh in my pandaboard in case you want to check configuration / change stuff & reboot should work14:22
guerbyrsalveti, ok thx :)14:22
rsalvetito see if that change affected this case14:22
guerbyI'll be mostly offline until this weekend14:22
rsalvetiguerby: yeah, saw the email, thanks for that :-)14:22
=== renato is now known as renatofilho
=== njpatel_ is now known as njpatel
rsalvetiGrueMaster: hey, can you later check bug 688765?15:55
ubot2Launchpad bug 688765 in linux (Ubuntu) "Can't init uart3 (no clocks available) at Beagleboard-xM (affects: 1) (heat: 8)" [Undecided,In progress] https://launchpad.net/bugs/68876515:55
rsalvetitime for food now15:55
GrueMasterSure, is there a patched kernel finally?15:55
rsalvetiGrueMaster: I created a different bug for our kernel, pointing the patches and a kernel deb file to test15:56
GrueMasterI thought we were on the mainstream kernel for omap?  Why file a different bug instead of posting a patch to the existing one?15:58
rsalvetiGrueMaster: the other bug was for linaro's kernel15:59
rsalvetiand this is for maverick15:59
GrueMasterAh.  I could have sworn there was a bug against the main kernel.  Will look into after coffee.16:00
Mark____Has anyone gotten the BeagleBoard Unbuntu install working at: https://wiki.ubuntu.com/ARM/BeagleNetInstall ?17:10
Mark____I think there are some important files missing from the instructions.17:10
GrueMasterMark____: I had tested this during Lucid timeframe and it worked fine.  I'll double check the instructions.  What issues are you seeing?17:11
GrueMasterBear in mind this is not tested on a BeagleXM (and likely won't work).17:12
Mark____I thought the FAT partition on the SD card needed a file named MLO17:12
Mark____which would load x-boot-loader.17:12
GrueMasterThe beagleboard should have that preinstalled in nand.17:13
Mark____I'm trying to boot on a Beagle xM and nothing is happening on the serial port.17:13
GrueMasterIf you power up the beagleboard with no USB or SD card, do you see anything on the serial console?17:13
GrueMasterAh, XM.  Not supported in Lucid.17:13
GrueMasterXM doesn't have nand, hence no boot loader.17:14
Mark____How should I get Ubuntu running on an xM?17:14
GrueMasterI'll see if I can get the instructions updated (or some workaround) for XM.  For now, use http://cdimage.ubuntu.com/ubuntu-netbook/ports/releases/10.10/release/ubuntu-netbook-10.10-preinstalled-netbook-armel+omap.img.gz and the instructions at https://wiki.ubuntu.com/ARM/OMAPMaverickInstall17:16
GrueMasterThere is an extra step involved if you have an XM rev A3 or later.17:17
GrueMasterThe instructions are there.17:17
Mark____Thanks.  I'll check it out now.17:18
GrueMasterThis will give you a netbook image, but once running it is fairly easy to customize.  We are working on a pre-installed minimal image for Natty, for developers that want something other than netbook installs.17:19
Mark____I'm doing a Beagle workshop for educators next month and I'm looking for some dirt-simple instructions they can give to they students17:22
Mark____and have something impressive running in a short time.17:23
Mark____I thought I had it with Ubuntu, but we aren't there yet.17:23
GrueMasterI haven't tested Maverick netboot on omap, but I'm sure some simple instructions could be cobbled together.17:23
GrueMasterUnless they have a hard requirement for no kvm support, the netbook image should work well for them.17:24
GrueMasterbtw:  I updated the net install instructions warning about the lack of support for XM.  Thanks for pointing it out.17:26
Mark____I guess I can pull out my C4 board and see if that works.17:28
Mark____It looks like it's working on the C4.17:33
Mark____Now I have to hook up the ethernet adapter.17:33
GrueMasterYea, netboot works a little better that way.  :P17:35
NCommanderMark____: netboot is kinda unsupported, it should work, but if it breaks, you get to keep both pieces17:48
Mark____Well....  The C4 is set up and it looks like it is doing something.17:50
Mark____If I get this SD working on the C4 can I switch to the xM and have it work?17:51
GrueMasterMark____: Try using the maverick netboot.  That would have better success, as it has kernel support for XM.  But you will need to create a 70-100M fat partition first.  And I am not sure if x-loader and u-boot would automatically install there.  It would take a little work, but it is doable.18:06
=== scottb-afk is now known as scottb
=== scottb is now known as sbambrough
jcrigbytgall_foo, ping?21:09
tgall_foojcrigby, pong21:14

Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!